There are two approaches to take regarding temperature changes and the effects they have on measurement. The first is to control the shop or lab environment to such an extent that, whatever temperature variations remain, they will have no significant effect. The alternative is to control and correct the key items involved-the gages and the product being measured. The simplest method for most folks is to control and correct.

Editor’s note: This is the second part of a two- part series.

In my last column I covered some of the basics regarding temperature changes and the effects they have on measurement. Fortunately, unlike the climate change folks, we don’t have to worry about our labs being flooded because of an icecap melting or polar bears in the parking lot. But we do have to deal with them.

There are two approaches to take. The first is to control the shop or lab environment to such an extent that, whatever temperature variations remain, they will have no significant effect. The alternative is to control and correct the key items involved-the gages and the product being measured. The simplest method for most folks is to control and correct. Here are some ways to do this:

First you have to know what you are dealing with in the way of temperature change, and this is done by recording it on a regular basis. This information could provide answers to problems down the road.

If the products, instruments and masters are steel, you only need concern yourself with differences between them rather than how close they are to 20 C.

Avoid taking measurements on components that are “hot off the machine” as there will be a dramatic difference in temperature, especially if they are brass, bronze or aluminum. If you can’t wait for these items to cool, then you have to deal with it. One way is to take a reading on a component and then let it cool so you can see what the effect will be dimensionally. You may have to shift your acceptable tolerance band upward so that even though the parts are at the wrong temperature during measurement, they’ll be correct at the right temperature.

Small components will not exhibit radical changes due to temperature variations. But if your tolerances on a diameter, for example, are less than ±0.001 inch, you may be able to proceed with caution, but your digital indicator will be changing its interpretation of “zero” as things cool.

A trick that is often used when components with a high CTE, such as brass, are involved but the setting masters are steel is to carefully measure a component at the beginning of the shift and record the readings. This component is tagged and becomes a secondary master-one that is checked from time to time through the shift. If the readings remain relatively stable, you can be assured that the ambient temperature is as well.

In my first column in this series I noted a basic that must be taken into account: rate of change, which, if ignored, could create problems. Let’s say you have a ½-inch or 12-millimeter diameter shaft you’ve decided will be measured using your coordinate measuring machine. Obviously, the effects of temperature change on the diameter measurements won’t be that great. But, if the shaft is 20 inches long and you have to measure that length, you could have a major problem, even if it is made of steel. Minor temperature variations will take a while to affect your CMM but not the shaft, which has very little metal compared to the machine.

You can buy electronic gages that monitor part and master temperature and compensate the readings you obtain but these can be costly. On the other hand, if the suggestions here don’t solve your temperature problems, they are worth the money.

There are other ways to deal with these problems, but like any problem-solving, first you must determine if you have a problem and its magnitude. Unlike the global warming climate change folks, we can’t blame someone or something else for the apparent problem without hard evidence to back us up. And we can’t count on consensus “science,” international politics or the UN for support.